78 A RESPIRATION CALORIMETER. 



jecting when necessary a considerable quantity of air (30 to 70 liters) 

 immediately after the beginning of the experimental period at 7 a. m., 

 thereby making use of the data obtained from the analysis of the air 

 at 7 a. m. in determining the composition and amount of the oxygen 

 and nitrogen rejected. 



Under these conditions the analysis of air at 7 a. m. holds good for 

 the first portion of air rejected, but, in order to keep up the normal 

 volume of air in the system and thus not draw the rubber diaphragms 

 on the pans down too tightly, it soon becomes necessary to admit oxygen 

 into the main air-pipe and thereby alter the composition. To delay 

 this step as much as possible, and so diminish its effect on the compo- 

 sition of the air rejected, we allow the temperature inside the chamber 

 to rise somewhat, as is its tendency when the subject is moving around 

 vigorously, as at this time of day, and thus utilize the expansion of the 

 air to keep the pans partially filled even when considerable air is being 

 rejected. Under these conditions, while it is probably true that the per- 

 centage of oxygen in the last portion of the air rejected is somewhat 

 higher than that in the first portion, we have customarily assumed 

 that the analysis of air at 7 o'clock represents so nearly the actual 

 composition of the air that no correction is necessary. 



In rejecting air, the petcock on the entrance pipe of the Elster 

 meter, through which the air sample is usually drawn, is closed, and the 

 petcock connecting the entrance pipe of the meter with the main air- 

 pipe at a point between the pans and the inlet for oxygen is opened. 

 The screw pinchcock on the T tube, through which the air sample for 

 oxygen analysis is taken, is opened, thus permitting free exit of the 

 air leaving the suction-pump. The air is rejected as soon as possible 

 after the end of the second residual analysis. Inasmuch as there is 

 but little resistance between the main air-pipe and the meter, the gas 

 passes through the meter very rapidly, and consequently 50 to 60 liters 

 of air can be rejected in about 15 minutes. The manometer indicates 

 a slightly diminished pressure, amounting to not far from 20 mm. of 

 water, which, together with the temperature of the meter, is read 

 when half the air is rejected. 



From the volume of air as measured by the meter, the temperature 

 of the meter, the manometer reading, and the barometric pressure, it 

 is possible to calculate exactly the volume and weight of oxygen and 

 nitrogen rejected. 



Occasionally, especially when a sudden fall in barometric pressure 

 has, by virtue of the expansion of the gases in the closed system, so 

 retarded the admission of oxygen that the percentage of oxygen has 

 fallen considerably below normal, it has been necessary to reject air 



